1. Field of the Invention
The present invention relates to methods of depositing a film of a metal chalcogenide and a method of preparing an improved field-effect transistor including film of a metal chalcogenide. More particularly, the present invention relates to a method of preparing an improved field-effect transistor including a film of a metal chalcogenide as the channel layer.
2. Description of the Prior Art
The ability to deposit high quality semiconducting, metallic and insulating thin films forms one of the important pillars of present-day solid-state electronics. A solar cell may include, for example, a thin n-type semiconductor layer (˜0.25 μm) deposited on a p-type substrate, with electrical contacts attached to each layer to collect the photocurrent. Light-emitting diodes (LED's) are typically comprised of a p-n bilayer, which under proper forward bias conditions emits light.
Thin-film field-effect transistors, denoted here as TFT's, include thin p- or n- type semiconducting channel layers, in which the conductivity is modulated by application of a bias voltage to a conducting gate layer that is separated from the channel by a thin insulating barrier. The electronic materials that comprise modem semiconducting devices have typically been silicon based, but can equally be considered from other families of materials, in some cases potentially offering advantages over the silicon-based technologies. The current invention relates to a new method of solution-depositing chalcogenide-based electronic materials in the form of high-quality thin films for application in electronic devices. Ability to deposit from solution is particularly attractive since it opens the opportunity to employ a number of low-cost, low-temperature, relatively quick techniques such as spin-coating, printing, stamping, and dipping. While the discussion presented below primarily refers to the application of these films in TFT's, this discussion is meant to be representative, and the electronic films so deposited could equally be used in other electronic devices as well.
Thin-film field-effect transistors (TFT's), are widely used as switching elements in electronic applications, most notably for logic and driver circuitry within processor and display applications. Presently, TFT's for many lower-end applications, including those employed in active matrix liquid crystal displays, are made using amorphous silicon as the semiconductor. Amorphous silicon provides a cheaper alternative to crystalline silicon—a necessary condition for reducing the cost of transistors for large area applications. Application of amorphous silicon is limited, however, to slower speed devices, since the mobility (˜10−1 cm2/V-sec) is approximately 15,000 times smaller than that of crystalline silicon.
In addition, although amorphous silicon is cheaper to deposit than crystalline silicon, deposition of amorphous silicon still requires costly processes such as plasma enhanced chemical vapor deposition. The search for alternative semiconductors (i.e., not silicon), for use in TFT's and other electronic devices is therefore being vigorously pursued.
If a semiconducting material could be identified which simultaneously provides higher mobility and low-cost processing at moderate/low temperatures, many new applications can be envisioned for these materials, including light, flexible, very large-area displays or electronics constructed entirely on plastic.
Recently, organic semiconductors have received considerable attention as potential replacements for inorganic counterparts in TFT's (see, for example, U.S. Pat. No. 5,347,144 assigned to Garnier et al., entitled “Thin-Layer Field Effect Transistor With MIS Structure Whose Insulator and Semiconductor Are Made of Organic Materials”) and LED's [S. E. Shaheen et al., “Organic Light-Emitting Diode with 20 lm/W Efficiency Using a Triphenyldiamine Side-Group Polymer as the Hole Transport Layer,” Appl. Phys. Lett. 74, 3212 (1999)].
Organic materials have the advantage of simple and low-temperature thin-film processing through inexpensive techniques such as spin coating, ink jet printing, thermal evaporation, or stamping. Over the last few years, the carrier mobilities of the organic channel layers in OTFTs (organic TFTs) have increased dramatically from <10−4 to ˜1cm2/V-sec (comparable to amorphous silicon) [see, for example, C. D. Dimitrakopoulos and D. J. Mascaro, “Organic thin-film transistors: A review of recent advances,” IBM J. Res. & Dev. 45, 11-27 (2001)].
While very promising with regard to processing, cost, and weight considerations, organic compounds generally have a number of disadvantages, including poor thermal and mechanical stability. In addition, while the electrical transport in organic materials has improved substantially over the last 15 years, the mobility is fundamentally limited by the weak van der Waals interactions between organic molecules (as opposed to the stronger covalent and ionic forces found in extended inorganic systems).
The inherent upper bound on electrical mobility translates to a cap on switching speeds and therefore on the types of applications that might employ the low-cost organic devices. Organic semiconductors are therefore primarily being considered for lower-end applications.
One approach to improving mobility/durability involves combining the processibility of organic materials with the desireable electrical transport and thermal/mechanical properties of inorganic semiconductors within hybrid systems [D. B. Mitzi et al., “Organic-Inorganic Electronics,” IBM J. Res. & Dev. 45, 29-45 (2001)]. Organic-inorganic hybrid films have recently been employed as the semiconductive element in electronic devices, including TFTs (see, for example, U.S. Pat. No. 6,180,956, assigned to Chondroudis et al., entitled “Thin-Film Transistors with Organic-Inorganic Hybrid Materials as Semiconducting Channels”) and LEDs (see, for example, U.S. Pat. No. 6,420,056, assigned to Chondroudis et al., entitled “Electroluminescent Device With Dye-Containing Organic-Inorganic Hybrid Materials as an Emitting Layer”).
Several simple techniques have been described for depositing crystalline organic-inorganic hybrid films, including multiple-source thermal evaporation, single source thermal ablation, and melt processing.
Solution-deposition techniques (e.g., spin coating, stamping, printing) have also received recent attention and are particularly attractive since they enable the quick and inexpensive deposition of the hybrids on a diverse array of substrates. TFT's based on a spin-coated semiconducting tin(II)-iodide-based hybrid have yielded mobilities as high as 1 cm2/V-sec (similar to the best organic-based devices prepared using vapor-phase deposition and amorphous silicon).
Melt-processing of the hybrid systems has improved the grain structure of the semiconducting films, thereby leading to higher mobilities of 2-3 cm2/V-sec [D. B. Mitzi et al., “Hybrid Field-Effect Transistors Based on a Low-Temperature Melt-Processed Channel Layer,” Adv. Mater. 14, 1772-1776 (2002)].
While very promising, current examples of hybrid semiconductors are based on an extended metal halide frameworks (e.g., metal chlorides, metal bromides, metal iodides, most commonly tin(II) iodide). Metal halides are relatively ionic in nature, thereby limiting the selection of possible semiconducting systems with potential for high mobility. In addition, the tin(II)-iodide-based systems in particular are highly air sensitive and all processing must by done under inert-amosphere conditions. Furthermore, while the tin(II)-iodide-based systems are p-type semiconductors, it is also desireable to find examples of n-type systems, to enable applications facilitated by complementary logic. So far none have been identified.
Another alternative to silicon-based, organic, and metal-halide-based hybrid semiconductors involves the use of metal chalcogenides (e.g., metal sulfides, metal selenides, metal tellurides) as semiconductive elements for use within TFT's and other electronic devices. Some of the earliest solar cells [D. C. Raynolds et al. “Photovoltaic Effect in Cadmium Sulfide,” Phy. Rev. 96, 533 (1954)] and TFTs [P. K. Weimer, “The TFT—A New Thin-Film Transistor,” Proc. IRE 50, 1462-1469 (1964)] were in fact based on metal chalcogenide active layers. There are numerous examples of metal chalcogenide systems that are potentially useful as semiconductive materials. Tin(IV) sulfide, SnS2, is one candidate that has generated substantial interest as a semiconducting material for solar cells, with n-type conductivity, an optical band gap of ˜2.1 eV and a reported mobility of 18 cm2/V-sec [G. Domingo et al., “Fundamental Optical Absorption in SnS2 and SnSe2,” Phys. Rev. 143, 536-541 (1966)].
These systems might be expected to yield higher mobility than the organic and metal-halide-based hybrids, as a result of the more covalent nature of the chalcogenides, and also provide additional opportunities for identifying n-type semiconductors.
Reported mobilities of metal chalcogenides, for example, include SnSe2 (27 cm2/V-sec/n-type) [G. Domingo et al., “Fundamental Optical Absorption in SnS2 and SnSe2,” Phys. Rev. 143, 536-541 (1966)], SnS2 (18 cm2/V-sec/n-type) [T. Shibata et al., “Electrical Characterization of 2H-SnS2 Single Crystals Synthesized by the Low Temperature Chemical Vapor Transport Method,” J. Phys. Chem. Solids 52, 551-553 (1991)], CdS (340 cm2/V-sec/n-type), CdSe (800 cm2/V-sec/n-type) [S. M. Sze, “Physics of Semiconductor Devices,” John Wiley & Sons, New York, 1981, p. 849], ZnSe (600 cm2/V-sec/n-type), and ZnTe (100 cm2/V-sec/p-type) [B. G. Streetman, “Solid State Electronic Devices,” Prentice-Hall, Inc., New Jersey, 1980, p. 443].
While the potential for higher mobility exists, the increased covalency of the extended metal chalcogenide systems also reduces their solubility and increases the melting temperature, rendering simple and low-cost thin-film deposition techniques for these systems a significant challenge.
A number of techniques have been proposed and employed for the deposition of chalcogenide-based films, including thermnal evaporation [A. Van Calster et.al., “Polycrystalline cadmium selenide films for thin film transistors,” J. Crystal Growth 86, 924-928 (1988)], chemical vapor deposition (CVD) [L. S. Price et al., “Atmospheric Pressure CVD of SnS and SnS2 on Glass,” Adv. Mater. 10, 222-225 (1998)], galvanic deposition [B. E. McCandless et al., “Galvanic Deposition of Cadmium Sulfide Thin Films,” Solar Energy Materials and Solar Cells 36, 369-379 (1995)], chemical bath deposition [F. Y. Gan et al., “Preparation of Thin-Film Transistors With Chemical Bath Deposited CdSe and CdS Thin Films,” IEEE Transactions on Electron Devices 49, 15-18 (2002)], and successive ionic layer adsorption and reaction (SILAR) [B. R. Sankapal et al., “Successive ionic layer adsorption and reaction (SILAR) method for the deposition of large area (˜10 cm2) tin disulfide (SnS2) thin films,” Mater. Res. Bull. 35, 2027-2035 (2001)].
However, these techniques are generally not amenable to low-cost, high-thoughput (fast) solution-based deposition techniques such as spin-coating, printing and stamping. What is required for application of these techniques is a truly soluble precursor for the chalcogenide and a suitable solvent. Given the potential for high electrical mobility in the metal chalcogenide systems and both n- and p-channel devices, if a convenient and rapid solution-based technique could be identified for their deposition, the field of low-cost solution-based electronics, currently primarily being pursued within the context of organic electronics, might be extended to higher-end applications, such as logic circuitry and very large area displays.
Spray pyrolysis is one technique employing the rapid decomposition of a soluble precursor [M. Krunks et al., “Composition of CulnS2 thin films prepared by spray pyrolysis,” Thin Solid Films 403-404, 71-75 (2002)]. The technique involves spraying a solution, which contains the chloride salts of the metal along with a source of the chalcogen (e.g., SC(NH2)2), onto a heated substrate (generally in the range 250-450° C.).
While metal chalcogenide films are formed using this technique, the films generally have substantial impurities of halogen, carbon or nitrogen. Annealing in reducing atmospheres of H2 or H2S at temperatures up to 450° C. can be used to reduce the level of impurities in the film, but these relatively aggressive treatments are not compatible with a wide range of substrate materials and/or require specialized equipment.
Ridley et al. [B. A. Ridley et al., “All-Inorganic Field Effect Transistors Frabricated by Printing,” Science 286, 746-749 (1999)] describes CdSe semiconducting films that are printed using a soluble metal chalcogenide precursor formed using organic derivatized CdSe nanocrystals. This technique, however, requires the formation of nanocrystals with tight control on particle size distribution in order to enable effective sintering during a postdeposition thermal treatment. The particle size control requires repeated dissolution and centrifugation steps in order to isolate a suitably uniform collection of nanocrystals.
Further, reported TFT devices prepared using this technique exhibited unusual features, including substantial device hysteresis and a negative resistance in the saturation regime, perhaps as a result of trap or interface states either within the semiconducting film or at the interface between the semiconductor and the insulator.
Dhingra et al. [S. Dhingra et al., “The use of soluble metalpolyselenide complexes as precursors to binary and ternary solid metal selenides,” Mat. Res. Soc. Symp. Proc. 180, 825-830 (1990)] have also demonstrated a soluble precursor for metal chalcogenides that can be used to spin coat films of the corresponding metal chalcogenide (after thermal treatment to decompose the precursor).
However, in this process, the species used to solublize the chalcogenide framework (i.e., quaternary ammonium or phosphonium polyselenides), which ultimately decompose from the sample during the heat treatment, are very bulky and most of the film disappears during the annealing sequence (e.g., 70-87 %). The resulting films consequently exhibit inferior connectivity and quality. The large percentage of the sample that is lost during the thermal treatment implies that only relatively thick films can be deposited using this technique, since thin films would be rendered discontinuous (the above mentioned study considered films with thickness ˜25-35 μm). Additionally, relatively high temperatures are required for the thermal decomposition of the polyselenides (˜530° C.), making this process incompatible with even the most thermally robust plastic substrates (e.g., Kapton sheet can withstand temperatures as high as 400° C.).
A study has also concluded that films of crystalline MoS2 can be spin coated from a solution of (NH4)2MoS4 in an organic diamine [J. Pütz and M. A. Aegerter, “Spin-Coating of MoS2 Thin Films,” Proc. of International Congress on Glass, vol. 18, San Francisco, Calif., Jul. 5-10, 1998, 1675-1680]. However, high-temperature post-deposition anneals are required to achieve crystalline films (600-800° C.), rendering the process incompatible with organic-based flexible substrate materials.
A similar procedure has led to the formation of amorphous As2S3 and As2Se3 films [G. C. Chern and I. Lauks, “Spin-Coated Amorphous Chalcogenide Films,” J. Applied Phys. 53, 6979-6982 (1982)], but attempts to deposit other main-group metal chalcogenides, such as Sb2S3 and GeSx have not been successful, due to the low solubility of the precursors in the diamine solvents [J. Pütz and M. A. Aegerter, “Spin-Coating of MoS2 Thin Films,” Proc. of International Congress on Glass, vol. 18, San Francisco, Calif., Jul. 5-10, 1998, 1675-1680].
Improvement in the above-described solution-based processes are therefore required for practical applications, especially for the fabrication of crystalline films of main-group metal chalcogenides, such as those derived from Ge, Sn, Pb, Sb, Bi, Ga, In, Tl, which can potentially provide high mobility in a semiconductor with reasonable band gap for a transistor.